[2A09-10] International sess.(3)

1:30 PM - 1:45 PM JST(4:30 AM - 4:45 AM UTC)Japanese

[2A10] Development of AI-Based Techniques for Extrapolating High-Boiling-Point Components in Petroleum Residual Fractions

○Yuya Murakami¹, Kotaro Matsumoto², Hiromasa Arai² (1. Shizuoka University, 2. Japan Petroleum and Carbon Neutral Fuels Energy Center)

Keywords:

FT-ICR MS,machine learning,Petroleomics

Heavy crude oil comprises a complex mixture of compounds, including extremely heavy species with boiling points near 1000 degC. Due to their structural complexity, these components are challenging to characterize using conventional analytical techniques, and their molecular structures remain largely unknown. This study presents an integrated approach that combines an auto-generation algorithm for candidate molecular structures with AI-based classification to identify unknown compounds present in crude oil.The results suggest the possible presence of high-molecular-weight polycyclic compounds containing more than 70 carbon atoms, which have been difficult to detect using traditional methods.